Deploy OpenShift AI and Elasticsearch vector database

We’ve provided you with a Jupyter notebook to ingest data into your Elasticsearch database. You’ll walk through setting up an OpenShift AI project with the necessary environment variables to access the Elasticsearch vector database from the provided Jupyter notebook.

In order to get the full benefit from taking this lesson, you need to:

• Install the OpenShift CLI.
• Have the appropriate access on your OpenShift cluster to install the OpenShift AI and Elasticsearch operators.
• Have the appropriate access on your OpenShift cluster to create projects.

In this lesson, you will:

• Set up OpenShift AI and Elasticsearch vector database.
• Run a notebook in OpenShift AI that ingests PDFs and URLs into a vector database.

OpenShift AI automated install

Follow the product documentation to install OpenShift AI or follow the instructions below:

1. Clone podman-ai-lab-to-rhoai.

2. Log in to your OpenShift cluster in a terminal with the API token. You can get your API token from the OpenShift web console in the top right corner of the page under admin. Click Copy login command (Figure 1).

   

   Your OpenShift CLI command should look similar to the one below:

   oc login --token=<YOUR_OPENSHIFT_API_TOKEN> --server=https://<YOUR_OPENSHIFT_API_URL>:6443

3. We'll first deploy the OpenShift AI operator:

   oc apply -k ./components/openshift-ai/operator/overlays/fast

4. Now we'll create a data science cluster. Make sure the operator is fully deployed before creating the data science cluster:

   oc get pods -n redhat-ods-operator --watch 
   NAME					            READY		STATUS
   Rhods-operator-5f7cf677cd-wp8vd 	1/1		    RUNNING

5. Once the pod has a Running status and is ready, you can run the below command:

   oc apply -k ./components/openshift-ai/instance/overlays/fast

Elasticsearch automated install

Follow the product documentation to install Elasticsearch or follow the instructions below:

1. Clone podman-ai-lab-to-rhoai.

2. We'll now deploy the Elasticsearch operator. This will be our vector database:

   oc apply -k ./components/elasticsearch/base/

3. Now we can create an Elasticsearch cluster instance. Make sure the Elasticsearch operator pod is in a Running state and ready:

   oc get pods -n elastic-vectordb --watch 
   NAME						        READY		STATUS
   Elastic-operator-5848f6968d-97qjj	1/1		    RUNNING
   
   oc apply -f ./components/elasticsearch/cluster/instance.yaml

Ingest data into the Elasticsearch vector database

Now that the Elasticsearch operator has been deployed and an instance has been created, we can ingest some data and query the vector database.

1. Go to your OpenShift AI Dashboard. You can get the URL either from the below oc command or the OpenShift web console:

   oc get routes -n redhat-ods-applications

   OR

   Select the redhat-ods-applications project, then Networking > Routes, then open the URL under Location (Figure 2).

   

2. Log in to the dashboard, and select Data Science Projects. Select Create data science project (Figure 3).

   

3. Name the project "podman-ai-lab-rag-project" and click the Create button (Figure 4).

   

4. We'll now create a workbench where we can upload a Jupyter notebook to ingest data into the Elasticsearch vector database. We'll then test it out by querying for some data. Select Create a workbench (Figure 5).

   

• Name the workbench "elastic-vectordb-workbench". Select a Standard Data Science notebook and a Medium size (Figure 6).

  

  You'll also want to set two environment variables so we can connect to Elasticsearch from the notebook. See below:

  • CONNECTION_STRING: Copy the CLUSTER-IP and PORT from this oc command:

    • oc get service elasticsearch-sample-es-http -n elastic-vectordb

    • Add the CONNECTION_STRING key/value as a ConfigMap environment variable.
       
  • PASSWORD: Create a secret environment variable with the Elasticsearch secret value:

    • oc get secret elasticsearch-sample-es-elastic-user -n elastic-vectordb -o jsonpath="{.data['elastic']}" | base64 -d > elastic_pass.txt

    • Add the PASSWORD key/value as a Secret environment variable. The password is in the elastic_pass.txt file that was created by the above oc command (Figure 7).

      

6. Click the Create Workbench button. Your workbench should start in a few minutes.
7. Open your workbench after it has started and log in.

8. Upload or import the ./notebooks/Langchain-ElasticSearchVector-ingest.ipynb notebook to your workbench (Figure 8).

   

9. Run the first 3 cells (Shift + Enter in each cell to run). Make sure you can connect to your Elasticsearch cluster (Figure 9).

   

   Note: You can safely ignore the pip dependency error during the package installations and you can ignore any TLS warnings when connecting to Elasticsearch

10. Continue to run through each cell while reading through what is occurring in each one. The "Create the index and ingest the documents" cell is where all of the websites and PDFs are stored with embeddings into the Elasticsearch vector database (Figure 10).

    

    This cell will take a while to run. Good time for a coffee break.

11. After all of the data is stored in our vector database, we can directly query it. Run the last 2 cells to make sure the data was stored successfully and that we get back the results we expect (Figure 11).

    

You now have OpenShift AI and Elasticsearch deployed. Data has been ingested into the Elasticsearch vector database and you were able to query data from the Jupyter notebook in OpenShift AI. Head over to the next lesson to deploy an LLM on OpenShift AI and update the chat recipe to connect the LLM and vector database to create an RAG chatbot!